TOPIC 2: PROBABIILITY DISTRIBUTIONS

2.1. Definitions
1. Random variable
-It is a variable whose numerical value is determined by the outcome of a
random trial.

-Types of random variables;

 Discrete random variable
 Continuous random variable

-Discrete random variables – refers to the random variables that takes only a countable
number of values e.g. number of heads in three coin tossing.

-Continuous random variables- -are variables that take any value in an interval of
numbers e.g. measured data on heights, temperature.

2. Probability distribution
- It is a function that permits us to compute the probability for any event that is defined
in terms of value of the random variable.

Note: The random variable is denoted by capital letters (X) while the value it assumes
by small letters (x)

2.1. Probability Distribution of a Discrete Random Variable

-The probability distribution for a discrete random variable X associates with each of
the distinct outcomes xi = (i= 1, 2,3,…k)

-The probability distribution of a discrete random variable X must satisfy the

following two conditions:
1. P(X = x) ≥ 0 for all values
2. ∑𝑎𝑙𝑙 𝑥 𝑃(𝑋 = 𝑥 ) = 1

Expected value and Variance

-The expected value of a discrete random variable X is denoted by E{𝑋} and defines as
E{𝑋} = ∑𝑘𝑖=1 𝑥𝑖 𝑝(𝑥𝑖 )
Where 𝑝(𝑥𝑖 ) = p (X=𝑥𝑖 )

-The variance of a discrete random variable is denoted by

𝜎 2 = V (X) =E (𝑋 − 𝜇)2 =∑𝑎𝑙𝑙 𝑥(𝑥 − 𝜇)2 . P(x)

2.1.1. Binomial Distribution

-It is a widely used distribution concerned with discrete random variables.
-It describes discrete data resulting from Bernoulli process.

The Bernoulli Process

-The characteristics of Bernoulli process are;

1
 i. Dichotomy-This means that each trial has only two mutually exclusive and
collectively exhaustive possible outcomes in the experiment (eg. Success or
failure)
ii. Stability – it means that the probability of an outcome of any trial is known and
remains fixed for all the trials.
iii. Independence-it means that the trials are statistically independent.

-Binomial random variable is a random variable that counts the number of successes
in many independent, identical Bernoulli trials.

-The probability distribution of Binomial Random Variable is called the Binomial

distribution.

Conditions for a Binomial model

i. A finite number ‘n’ trials are carried out
ii. The trials are independent
iii. The outcome of each trial is deemed either a success or failure
iv. The probability 𝑝 of a successful outcome is the same for each trial.

2.1.1.1. BINOMIAL PROBABILITY FUNCTION

-The distribution of Binomial random variable is described by two parameters, n and p
X~ B (n , p)

This distribution indicates that X is Binomially distributed with n number of independent

trials and p probability of success is each trial. The letter B stands for binomial.

Note: The probability of failure is given by

𝑞 = 1- 𝑝
-If X~ B (n, p), the probability of obtaining 𝑥 successes in 𝑛 trials is given by;
P(X=x) =𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥 for x=0, 1, 2, 3…n

-The probability distribution satisfies two essential properties of probability distribution;

(i)= f(x) ≥ 0
(ii) ∑ 𝑓(𝑥 ) = 1

2.1.1.2. Mean and variance of Binomial Distribution

(i) The mean of binomial distribution is given as;
𝐸(𝑋) = 𝑛𝑝
(ii) The variance of the binomial distribution is given as
𝜎 2 = 𝑛𝑝𝑞

Example 1
-The mean of a binomial distribution is 40 and standard deviation 6. Calculate n, p and
q.

Solution
Mean = E(X) =𝜇 = np =40
Variance=𝜎 2 = 𝑛𝑝𝑞

2
Standard deviation = √𝑛𝑝𝑞 =6, so Variance = npq = 36
36
Therefore, np= 40 and npq =36, so q= 40 = 0.9
P= 1 - q, so q= 1- 0.9 = 0.1
40 40
np=40 but p =0.1 , so n= 𝑝 = 0.1 = 400
Therefore n=400, p= 0.1, q= 0.9

Example 2
The incidence of occupational disease in an industry is such that the workers have a
20% chance of suffering from it. What is the probability that out of six workers 4 or
more will contract the disease?

Solution
20
-The probability of a worker who is suffering from the disease p= 100 =0.2
q=1-p, q= 1-0.2=0.8
-The probability of 4 or more i.e. 4, 5, or 6 will contract the disease is given by
p(X≥4)= p(x=4) + p(x=5) + p(x=6)
p(X≥4)= 6𝐶4 0.24 0.86−4 + 6𝐶5 0.25 0.86−5 + 6𝐶6 0.26 0.86−6
=0.01536 + 0.001536 + 0.000064
=0.01696

Example 2
Assume that on an average one telephone number, out of fifteen is busy. What is the
probability that if six randomly selected telephone numbers are called?
a) Not more than three will be busy
b) At least three of them will be busy

Solution
1 14
p=15=0.0667; q=15=0.93333; n=6
a) P(X≤ 3)=p(x=0) + p(x=1) + p(x=2) +p(x=3)
= 6𝐶0 0.06670 0.93336−0 +6𝐶1 0.06671 0.93336−1 +6𝐶2 0.06672 0.93336−2
+6𝐶3 0.06673 0.93336−3

=0.6609 + 0.2834 +0.05063+0.004824

=0.9998
b) Probability that at least three telephone numbers are busy is given by;
P(X≥3) =p(x=3) + p(x=4) + p(x=5) + p(x=6)
= 6𝐶3 0.06673 0.93336−3 + 6𝐶4 0.06674 0.93336−4 +
65 0.06675 0.93336−5 + 66 0.06676 0.93336−6

= 0.004824 + 0.0002586 + 0.000007392+0.00000008806

=0.0051
Activity
i. In Naivas supermarket, 70% of customer pay by credit card. Find the probability
that in a randomly selected samples of 12 customers
a. Five pay by credit card
b. AT least 2 pay by credit card

3
 c. Less than 2 pay by credit card
d. Find the mean and variance described by this situation

2.1.1.3. Fitting a Binomial Distribution to Practical data

Example 1
-A biased coin is tossed four times and the number of heads noted. The experiment is
performed 500 times and the results are summarized in the table below

Number of heads 0 1 2 3 4

Frequency 12 50 151 200 87

Required:
a) From the experimental data estimate the probability of obtaining a head when
the coin is tossed.
b) Using binomial distribution with the same mean, calculate the theoretical
probabilities of obtaining 0, 1, 2, 3 and 4 heads.

Solution
No of Frequency Xf Theoretical probabilities Expected
heads (X) (f) probabilities
𝑃 (𝑋 = 𝑥 ) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥

0 12 0 4𝐶0 0.650 0.354−0 =0.015 0.015 ×

500= 8

1 50 50 4𝐶1 0.651 0.354−1 =0.1115 0.1115×

500 = 𝟓𝟔

2 151 302 4𝐶2 0.652 0.354−2 =0.3105 0.3105 ×

500 =155

3 200 600 4𝐶3 0.653 0.354−3 =0.3844 0.3844 ×

500=192

4 87 348 4𝐶4 0.654 0.354−4 =0.1785 0.1785×

500 =89
∑ 𝒇 = 500 ∑ 𝒇𝒙= 1300

Mean =𝑿 ̅ = 𝟏𝟑𝟎𝟎 = 2.6.

𝟓𝟎𝟎
a) E(X) = np , but np =2.6, n=4
𝟐.𝟔
P= 𝟒 = 0.65

Activity
The screws produced by a certain machine were checked by examining the number of
defectives in a sample. The following table shows the distribution of 128 samples
according to the number of defective items they contained;

4
 No. of 0 1 2 3 4 5 6 7
defectives
No. of 7 6 19 35 30 23 7 1
samples

2.1.2. Poisson Distribution

-It is named after the French mathematician, Poisson who published its derivation in
1837.

2.1.2.1. Conditions of Poisson distribution

i. The occurrence of the events is independent.
ii. 𝛾, the mean number of occurrences in a given interval is known ad finite
iii. Theoretically, an infinite number of occurrences of the event must be possible
in the interval.

-If the above conditions are satisfied, X is said to follow Poisson distribution written
as;
X~ P0 (𝛾)

-The Poisson distribution is given by;

𝑒 −𝛾 𝛾 𝑥
f(x)=P(X=x) = for x=0,1,2,3,…,
𝑥!

where 𝑒=a constant whose value =2.7183

𝛾=mean or average number of occurrences of random event
x=number of occurrences of the random event

-The Poisson distribution satisfies the two essential properties of a probability

distribution, i.e.
(i)= f(x) ≥ 0
(ii) ∑ 𝑓 (𝑥 ) = 1

-The Poisson distribution is completely defined by the parameter,𝛾.

2.1.2.2. Mean and Variance of the Poisson Distribution

a. The mean, E(X),of the Poisson distribution is 𝛾
b. The variance, Var(x),of the Poisson distribution is given by 𝜎 2 =𝛾

Note: The Poisson distribution may be used to approximate binomial distribution when
𝑛 is large and 𝑝 is small and therefore, is regarded as the limit of the binomial
distribution.

Example 1
On the average, one in 400 items is defective. If the items are packed in boxes of 100,
what is the probability that any given box of items will contain;
(i) No defective
(ii) Less than two defectives

5
 (iii) One or more defectives
(iv) More than three defectives

Solution
1
P=400 n=400
100
E(x)=𝛾 =np=400 = 0.25

𝑒 −0.25 0.250
(i) P(x=0)= =0.7788
0!

(ii) P(x≤ 1)= p(x=0) + p(x=1)

𝑒 −0.25 0.250
= = 0.7788
0!
𝑒 −0.25 0.251
= =0.1947
1!
=0.7788+0.1947
=0.9735

(iii) P(x≥1)= 1- P(x≥1)=1- 0.7788=0.2212

(iv) P(x≥4)=1- [p(x=0) +p(x=1) +p(x=2) + p(x=3)

= 1- (0.7788 + 0.1947 + 0.02434+0.002029)
= 1-0.99984
=0.00014

Example 2
A factory produces blades in packets of 10. The probability of a blade to be defective is
0.2%. Find the number of packets having two defective blades in a consignment of
10000 packets.

Solution
E(x=) =𝛾=np=10× 0.002 = 0.02
𝑒 −0.02 0.022 0.9802×0.0004
P(x=2) = = = 0.000196
2! 2!

Total number of packets = 10000 × 0.000196=1.96

Assignment
1.A student finds that the average number of amoeba in a 10 ml of cold water from a
particular pond is 4. Assume that the number of amoeba follow Poisson distribution,
find the probability that in a 10 ml sample of such water.
a. There are more than 5 amoeba
b. There are exactly 3 amoeba
c. There are at most 4 amoeba

2.1.2.3. Fitting Poisson distribution to Practical data

Example 1
The following table gives the number of days in a 50-day period during which
automobile accidents occurred in a city. Fit the Poisson distribution of the data.

6
 Number of accidents 0 1 2 3 4

Number of days 21 18 7 3 1

Solution
x f fx Theoretical Probabilities Expected
probabilities

0 21 0 P(x=0)=
𝑒 −0.9 0.90
=0.4066 0.4066 ×
0!
50=20 days

1 18 18 P(x=1)=
𝑒 −0.9 0.91
=0.3659 0.3659×
1!
50=18 days

2 7 14 P(x=2)=
𝑒 −0.9 0.92
=0.1647 0.1647 × 50=8
2!
days

3 3 9 P(x=3)=
𝑒 −0.9 0.93
=0.0494 0.0494 ×50=3
3!
days

4 1 4 P(x=4)=
𝑒 −0.9 0.94
=0.0111 0.0111 ×50=1
4!
day
∑ 𝑓=50 ∑ 𝑓𝑥 = 45

∑ 𝑓𝑥 45
Mean =𝑋̅ = ∑ 𝑓 =50 = 0.9

Assignment
Five hundred Tv sets are inspected as they come off the production line and the number
of defects per set recorded below;
Number of defects (x) 0 1 2 3 4

Number of sets 368 72 52 7 1

Fit the Poisson distribution of the data.

2.1.3. Geometric Distribution

-For an event to be described as using a geometric model;
i. independent trials re carried out
ii. The outcome of each trial is deemed either a success or failure
iii.The probability (p) of a successful outcome is the same for each trial
iv.The discrete random variable x is the number of trials needed to obtain the
first successful outcome

-If the above condition ns are satisfied x is said to follow a geometric distribution;
X ~ Geo (p)

7
-If X ~ Geo (p) , the probability that the first success is obtained at 𝑟 𝑡ℎ attempt is given
by;
P(X=r)=𝑞 𝑟−1 .p ; r=1, 2,…

2.1.3.1 Mean and Variance of Geometric Distribution

a. The mean of geometric distribution is given by
1
E(X) = 𝑝
b. The variance of geometric distribution is given by
𝑞
Var(x) =𝑝2

Example 1
Jack is playing a ball game in which he needs to throw a 6 with an ordinary die in order
to start the game. Find the probability that
a. Exactly 4 attempts are needed to obtain a 6
b. At least 2 attempts are needed to obtain a 6
c. He is successful in throwing a 6 in three or fewer attempts
d. He needed more than 3 attempts to obtain a 6

Solution
1 1 5
p=6 q= 1-6 =6

1 54−1
a. P(r=4) =𝑞 𝑟−1 .p=6 × =0.0965
6
b. P(r≥2)=1- p(r <2)
=1- p(r=1)
1 51−1
=1- (6 × )
6
1
=1- 6= 0.83333
c. P(r≤3)= p(r=1) + p(r=2) + p(r=3)
1 51−1 1
=p(r=1)= 6 × =6 =0.1667
6
1 52−1 5
=p(r=2)= 6 × =36 =0.13889
6
1 53−1 25
=p(r=3) =6 × =216 = 0.1157
6
0.4213
Activity
Attempt part d

2.2. Probability Distribution of a Continuous Random Variable

-A probability function for a continuous random variable is called a continuous
probability function since the domain of the function is continuous.

-For a continuous random variable, the corresponding function 𝑓(𝑥) is called a

probability density function (p.d.f).

2.2.1. Normal Distribution

-It was discovered by De Moivre in 1733

8
-A random variable X is said to have a normal distribution with parameters 𝜇 (mean)
and 𝜎 2 (variance) if its density function is given by;
−1 𝑥−𝜇 2
1
y=f(x)=𝜎 𝑒 2( 𝜎
)
√ 2𝜋
Where;
y=the computed height of an ordinate at a distance of X from the mean
𝜎 = Standard deviation of the given normal distribution
𝜋 =the constant =3.1416
e=the constant =2.7183
𝜇 =mean of the given normal distribution

-Symbolically, it can be expressed as;

X~N(𝜇, 𝜎)

2.2.2.1. Features/Characteristics of Normal Distribution

i. It is bell shaped
ii. It is symmetrical about the 𝜇
iii.It extends from minus infinite to positive infinite, −∞ ≤ x ≤ ∞
iv.The total area under the curve is 1

Finding Probabilities
-The value of x must be standardized in order to use the standard normal tables
𝑋−𝜇
𝑍= 𝜎
- The standardized variable Z is called a standard normal variate which has mean zero
and standard deviation of one.

-In symbols Z ~ N(0,1)

Example 1
A random variable X is normally distributed with mean of 50 and standard deviation
of 100. Compute the probability 45≤ X ≤ 62.

Solution
𝑋−𝜇
𝑍= 𝜎
45−50 62−50
=p( 10
)≤Z≤( 10
)

=p(-0.5 ≤ Z ≤ 1.2)

z=-0.5 0 z=1.2

=0.3849 + 0.1915
=0.5764

9
Example 2
An industry produces an item whose weights are normally distributed with the mean 𝜇
=18.5kg and 𝜎 =1.5kg. What is the probability that an item chosen at a random weigh
more than 21.5kg?

Solution
P(X > 21.5) 𝜇 =18.5 𝜎 =1.5

𝑋−𝜇
𝑍=
𝜎

21.5−18.5
P(Z > )
1.5

P (Z > 2)

0 z=2
P(Z>2) =0.5- 0.4772

=0.0228

Example 3
Suppose that X has normal distribution with 𝜇 =10 and 𝜎 2 =4. Find
i. P ( -3≤ X ≤ 12)
ii. P(|𝑋| ≤ 5

Solution
i. P ( -3≤ X ≤12)
𝑋−𝜇
𝑍= 𝜎

−3−10 12−10
=p( )≤Z ≤( )
2 2

=P(-6.5 ≤ Z ≤ 1)

z=-6.5 0 z=1

=0.5 + 0.3413
=0.8413

ii. P (-5 ≤ X ≤ 5)

10
 𝑋−𝜇
𝑍= 𝜎
−5−10 5−10
= p( )≤z ≤( )
2 2

=p( -7.5 ≤ z ≤ -2.5)

z=-7.5 z=-2.5 0

=0.5- 0.4938
=0.0063

Example 4
If X is a random variable which is normally distributed with a mean 𝜇 and 𝜎 =𝜇 and if
p(X ≤8) =0.95. Determine the probability 4≤ X ≤ 11.

Solution
P(X ≤ 8) = 0.95

8−𝜇
P(Z≤ = 0.95
𝜇

From standard normal table 0.95 = ordinate value 1.65

8−𝜇
So =1.65
𝜇
=8-𝜇=1.65𝜇

8=2.65𝜇

𝜇=3

Therefore,
For P(4≤ X ≤ 11)
4−3 11−3
= p( 3 ) ≤ z ≤ ( 3 )

=p(0.3≤ z ≤ 2.7)

0 z=0.3 z=2.7

= 0.5-0.4965
11
=0.0035

Assignment
1. The time taken by the milk man to deliver milk to some busy streets is normally
distributed with a mean of 12 min and standard deviation of 2 min. He delivers
milk every day. Estimate the number of days during the year when he takes;
a. Longer than 17 min
b. Less than 10 min
c. Between 9 and 13 min
2. The heights of female students of a particular college are normally distributed
with a mean of 169cm and standard deviation of 9cm
a. Given that 80% of these female students have a height less than h. Find the
value of h.

2.3. The Normal Approximation to Binomial Distribution

-For the discrete random variable X, distributed binomially X ~ B(n, p), the 𝜇 =E(x)=np
and variance =𝜎 2 =npq. When n is large and mean not too far from 0.5, a normal
distribution with mean np and variance npq can also be used as an approximation to
binomial distribution.

-The rule is if X~ B(n,p) , and n and p are such that np>5 and nq > 5, then X ~ N
(np,npq) approximately.
-Therefore, binomial distribution approaches a normal distribution with standardized
𝑋−𝑛𝑝
normal variable Z= 𝑛𝑝𝑞 ~ N(0,1)
√

Example
Find the probability of obtaining 4, 5, 6 or 7 heads when a fair coin is tossed 12 times.
a. Using binomial distribution
b. Using normal distribution to the binomial distribution

Solution
a. P(X=x)= 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥

N=12, x=4, 5, 6,7

1 1
P=2 , q= 2

=p(x=4) + p(x=5) + p(x=6) + p(x=7)

=12𝐶4 0.54 0.512−4 +12𝐶5 0.55 0.512−5 + 126 0.56 0.512−6 + 12𝐶7 0.57 0.512−7

= 0.1208 + 0.1934 + 0.2256 + 0.1934

=0.7332

b. Check for
np=12 × 0.5= 6 >5

12
 nq=12 × 0.5= 6>5
npq= 0.5 × 12 × 0.5 = 3

X~ N(6,3)

P(4< X<7)-convert to continuity function

So you get, P( 3.5 < X < 7.5)

3.5−6 7.5−6
P( )< Z < ( )
√3 √3

=p( -1.44 <z< 0.87)

Z=-1.44 z=0.87

=0.4251 + 0.3078

=0.7329

2.4. The Normal Approximation to Poisson Distribution

If X follows a Poisson distribution with parameter 𝛾 , E(X) =𝛾 , var(X) =𝛾.
When 𝛾 is large say (𝛾 > 15) the normal distribution can be used to approximate X
such that X ~ N (𝛾, 𝛾 )

-Therefore, Poisson distribution approached normal distribution with standardized

𝑋−𝑚
normal variable i.e. Z= 𝑚 ~ N (0, 1)
√

Example
A radioactive disintegration give counts that follow a Poisson distribution with mean
count of 25 per second. Find the probability that the count is between 23 and 27
inclusive.

Solution
𝛾 = 25 𝜎=5
We are looking for the P(23 ≤ X ≤ 27) as a continuous function, it will be
P(22.5 ≤ X ≤ 27.5)

22.5−25 27.5−25
=P( )< Z < ( )
5 5

=p( -0.5 ≤ Z ≤ 0.5)

13
 z=-0.5 0 z=0.5

=0.1915 + 0.1915

=0.3830

14